INDEX

Explanations

summary or conclusion

np_acts-logits-general · gemini-2.5-flash-lite

This neuron spots the initial content words that begin quoted speech (i.e. the first few words inside quotation marks).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

quoted declarative statements, especially ones starting with first-person or other deictic pronouns.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

direct quotations that begin with first-person pronouns, particularly "I" and "We".

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

또한

0.96

 גם

0.90

 또한

0.80

 देखील

0.77

 ALSO

0.76

 anche

0.76

 কথাও

0.75

also

0.75

גם

0.73

$)$.

0.72

POSITIVE LOGITS

1.13

 definitely

0.91

absolutely

0.84

 absolutely

0.79

,’’

0.77

Definitely

0.77

,"

0.76

 Definitely

0.75

 ওঁ

0.75

 Absolutely

0.74

Activations Density 0.181%

summary or conclusion

This neuron spots the initial content words that begin quoted speech (i.e. the first few words inside quotation marks).

quoted declarative statements, especially ones starting with first-person or other deictic pronouns.

direct quotations that begin with first-person pronouns, particularly "I" and "We".

No Comments

No Known Activations

summary or conclusion

This neuron spots the initial content words that begin quoted speech (i.e. the first few words inside quotation marks).

quoted declarative statements, especially ones starting with first-person or other deictic pronouns.

direct quotations that begin with first-person pronouns, particularly "I" and "We".

No Comments

No Known Activations